The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses various shell commands to manipulate the local git repository, including git checkout, git stash, and git apply to reimplement branch logic.
[COMMAND_EXECUTION]: The instructions recommend using git commit --no-verify during the reimplementation process. This flag explicitly bypasses local pre-commit hooks, which may include security scanners, secrets detection tools, or linting checks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local repository's history and diffs, creating a potential attack surface if those files contain malicious instructions.
Ingestion points: Data is pulled into the agent's context via git log main..HEAD --oneline and git diff main...HEAD --stat in the SKILL.md workflows.
Boundary markers: The instructions do not define specific delimiters or "ignore" instructions to isolate the ingested diff content from the agent's primary logic.
Capability inventory: The skill can modify the local filesystem (via git commands) and interact with external services (via the /pr tool).
Sanitization: There is no evidence of content sanitization or validation for the data retrieved from git diffs before it is analyzed by the agent.

reimplement-branch